If the earth is swinging around the sun once a year in a mighty path or orbit, there must be changes of its position in space comparable in size with the orbit itself. And the stars ought to shift their apparent places on the sky to correspond with these changes in the terrestrial observer's position. The phenomenon is analogous to what occurs when we look out of a room, first through one window, and then through another. Any object on the opposite side of the street will be seen in a changed direction, on account of the observer's having shifted his position from one window to the other. If the object seemed to be due north when seen from the first window, it will, perhaps, appear a little east of north from the other. But this change of direction will be comparatively small, if the object under observation is very far away, in comparison with the distance between the two windows. This is what occurs with the stars. The earth's orbit, vast as it is, shrinks into almost absolute insignificance when compared with the profound distances by which we are sundered from even the nearest fixed stars. Consequently, the shifting of their positions is also very small--so small as to be near the extreme limit separating that which is measurable from that which is beyond human ken. Photography lends itself most readily to a study of this matter. Suppose a certain star is suspected of "having a parallax." In other words, we have reason to believe it near enough to admit of a successful measurement of distance. Perhaps it is a very bright star; and, other things being equal, it is probably fair to assume that brightness signifies nearness. And astronomers have certain other indications of proximity that guide them in the selection of proper objects for investigation, though such evidence, of course, never takes the place of actual measurement. The star under examination is sure to have near it on the sky a number of stars so very small that we may safely take them to be immeasurably far away. The parallax star is among them, but not of them. We see it projected upon the background of the heavens, though it may in reality be quite near us, astronomically speaking. If this is really so, and the star, therefore, subject to the slight parallactic shifting already mentioned, we can detect it by noting the suspected star's position among the surrounding small stars. For these, being immeasurably remote, will remain unchanged, within the limits of our powers of observation, and thus serve as points of reference for marking the apparent shifting of the brighter star we are actually considering.